A neuroscientist is challenging the idea that your brain makes decisions at all

A cognitive neuroscientist at Indiana University has published an argument that the standard model of how people choose between options may rest on a category mistake. Tom James, a professor in the university’s Department of Psychological and Brain Sciences, contends that a distinct decision-making stage has never actually been located inside the brain, because no dedicated neural circuit for it has ever turned up. What looks like deliberation followed by a choice, he argues, is better explained as one continuous process spread across sensing and moving. The paper does not claim that choices are fake; it claims that neuroscience has been searching for brain activity behind something that was never a separate physical step to begin with.

The argument appears in a paper titled “Sensorimotor Mechanisms of Decisions and Actions,” published in the Journal of Cognitive Neuroscience, and is built on analogies, robotics and a philosophical framework rather than new brain scans. James is not the first researcher to question how decisions are studied, but the paper takes direct aim at an assumption built into decades of experimental design: that a distinct decision-making stage exists between sensing the world and acting on it, waiting to be found.

The “Sandwich Model” That James Says Doesn’t Hold Up

For decades, both formal theory and everyday intuition have treated a decision as the filling between two slices of bread: perception on one side, action on the other, with a distinct thinking stage sandwiched in between. Researchers have identified real circuits for the outer layers — sensory systems that register a stimulus and motor systems that carry out a movement. What has never turned up, James argues, is a comparable circuit that performs the middle step. Much of model-based cognitive neuroscience is built around the assumption that such a stage exists, which he says reinforces the idea rather than testing it. The model also matches ordinary experience, since people naturally describe their own behavior as caused by weighing beliefs and desires before acting.

Swapping “Decision” for “Action Selection”

In place of a central decision stage, James proposes what he calls action selection: behavior that arises from the combined activity of sensory, sensorimotor and motor processes rather than a single step that directs them. He is not arguing that deliberation is imaginary. “Our actions feel like they are caused by decisions based on desires, beliefs, and intentions,” James said, according to Indiana University’s psychology department, adding that the language is useful for describing behavior even without a matching brain process behind it. The distinction he draws is between decisions as a way of talking about outcomes after the fact and decisions as literal, physical events located somewhere inside the skull.

A Wall-Following Robot With No Decisions Built In

To make the case concrete, James points to a simple robot equipped with only sensory, motor and sensorimotor components — no memory, no planning and no internal representation of goals. Given a basic rule to move toward targets while avoiding collisions, the robot ends up tracing the walls of a room, a behavior that looks deliberate and strategic from the outside. “It looks intentional. It looks strategic. It looks like the robot is making decisions. And yet, it is not,” James said, noting that nothing in the robot’s design performs anything resembling a choice between options. If simple hardware can produce the appearance of decision-making without any decision-making mechanism inside it, James argues, the same possibility has to be taken seriously for brains.

Borrowing Dennett’s Center of Mass to Explain Why a Decision Can’t Push Anything

James’s framework leans on ideas associated with philosopher Daniel Dennett, under which only physical events can cause other physical events. A decision, on this account, resembles a center of mass: a real and useful concept for describing an object’s behavior, but not itself a physical thing that can be relocated — moving it requires moving the mass it describes. Applied to the brain, the argument is that a decision cannot directly trigger a muscle movement, because a decision is not made of neurons and synapses the way sensory and motor activity is. James extends the same logic to an analogy about institutions, noting that saying “the university took certain actions” is a convenient shorthand for meetings, phone calls and individual choices, not a description of what a university literally is or does.

The Infinite-Regress Problem Dennett Called the Cartesian Theatre

James also revives Dennett’s objection to picturing the brain as run by an internal controller, sometimes called the Cartesian Theatre, after an imagined inner viewer who watches sensory input arrive and issues commands in response. “That person inside your brain would need another person inside its brain, which would need a person inside its brain and so on, in an infinite regress,” James said. “So the problem is never solved. It’s just passed on.” Proposing a controller, in other words, does not explain how decisions happen; it relocates the unexplained part one level deeper into the system, without ever accounting for how that inner controller does its own deciding.

Why the Paper Stops Short of New Brain Data

The paper is theoretical, built from argument and analogy rather than new scans or experiments, and James has acknowledged that the framework needs to be tested against real neural and behavioral data before it can replace the sandwich model it criticizes. He suggests studying choices in dynamic, real-world settings, rather than the controlled, single-trial tasks common in laboratories, since those tasks are partly designed around the very assumption he is questioning. His own lab has begun exploring the idea through embodied cognition and ecological psychology, fields that treat behavior as inseparable from the body and its immediate surroundings rather than as commands issued by an isolated brain region. The full study appears in the Journal of Cognitive Neuroscience, and its implications were summarized this week by ScienceAlert, which called the framework a genuine challenge to how cognitive neuroscience models behavior.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview